Portfolio optimization using elliptic entropy and semi-entropy of coherent fuzzy numbers

Gupta, Pankaj

doi:10.1016/j.ins.2022.09.032

Cited by 10 publications

(1 citation statement)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When the market is in a turbulent phase, the entropy of equity market indexes decreases, so the indexes that earn will be more predictable and regular. In his study of elliptic and hyperbolic pseudo-entropies, Gupta (2022) investigates the concept of elliptic entropy. He uses adaptive indexes and oval and semi-entropic measures to capture investor attitudes.…”

Section: Iet and Asset Valuationmentioning

confidence: 99%

Information Entropy Theory and Asset Valuation: A Literature Survey

Chortane,

Naoui

2022

IJABF

View full text Add to dashboard Cite

The purpose of this study is to review the empirical work applied to market efficiency, portfolio selection and asset valuation, focusing on the presentation of the comprehensive theoretical framework of Information Entropy Theory (IET). In addition, we examine how entropy addresses the shortcomings of traditional models for valuing financial assets, including the market efficiency hypothesis, the capital asset pricing model (CAPM), and the Black and Scholes option pricing model. We thoroughly reviewed the literature from 1948 to 2022 to achieve our objectives, including well-known asset pricing models and prominent research on information entropy theory. Our results show that portfolio managers are particularly attracted to valuations and strive to achieve maximum returns with minimal risk. The entropy-based portfolio selection model outperforms the standard model when return distributions are non-Gaussian, providing more comprehensive information about asset and distribution probabilities while emphasising the diversification principle. This distribution is then linked to the entropic interpretation of the no-arbitrage principle, especially when extreme fluctuations are considered, making it preferable to the Gaussian distribution for asset valuation. This study draws important conclusions from its extensive analysis. First, entropy better captures diversification effects than variance, as entropy measures diversification effects more generically than variance. Second, mutual information and conditional entropy provide reasonable estimates of systematic and specific risk in the linear equilibrium model. Third, entropy can be used to model non-linear dependencies in stock return time series, outperforming beta in predictability. Finally, information entropy theory is strengthened by empirical validation and alignment with financial views. Our findings enhance the understanding of market efficiency, portfolio selection and asset pricing for investors and decision-makers. Using Information Entropy Theory as a theoretical framework, this study sheds new light on its effectiveness in resolving some of the limitations in traditional asset valuation models, generating valuable insights into the theoretical framework of the theory.

show abstract

Section: Iet and Asset Valuationmentioning

confidence: 99%